| Message ID | 20200221090053.9982-1-hui.wang@canonical.com |
|---|---|
| Headers | show |
| Series | Fix hang issue if sof fails to load firmware | expand |
On 21.2.2020 11.00, Hui Wang wrote: > BugLink: http://bugs.launchpad.net/bugs/1861837 > > If the sof driver fails to load the firmware, it doesn't release the > resource, then it make the system hang when suspending and resuming. > > After this patchset is applied, please also apply the patchset we > just reverted 2 weeks ago, since this patchset is the fixing patches > for the patches we reverted (#1857409). If need me to resend the > reverted patches, I will resend them. > > > [Impact] > The machine hangs during suspend/resume if the sof driver fails to > load firmware. > > [Fix] > This bug is fixed in the mainline kernel v5.6-rc1, now backported > those patches to ubuntu kernel > > [Test Case] > sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot > then suspend and resume, the system will not hang anymore > > And if the sof driver load the firmware successfully, the audio > functions all worked well as before. > > [Regression Risk] > Low, the patches are backported from upstream kernel, and I tested > these patches on Lenovo and Dell dmic machines. > > > Kai Vehmanen (2): > ASoC: hdac_hda: Fix error in driver removal after failed probe > ASoC: SOF: trace: fix unconditional free in trace release > > Pierre-Louis Bossart (2): > ASoC: SOF: core: free trace on errors > ASoC: SOF: core: release resources on errors in probe_continue > > Ranjani Sridharan (1): > ASoC: SOF: Introduce state machine for FW boot > > sound/soc/codecs/hdac_hda.c | 4 +- > sound/soc/sof/core.c | 86 ++++++++++++++++++++-------- > sound/soc/sof/intel/hda-loader-skl.c | 1 - > sound/soc/sof/intel/hda-loader.c | 1 - > sound/soc/sof/intel/hda.c | 4 +- > sound/soc/sof/ipc.c | 17 ++---- > sound/soc/sof/loader.c | 19 ++++-- > sound/soc/sof/pm.c | 25 +++++++- > sound/soc/sof/sof-priv.h | 11 +++- > sound/soc/sof/trace.c | 7 ++- > 10 files changed, 125 insertions(+), 50 deletions(-) applied to osp1 oem-next, thanks
On 2020/2/21 下午7:23, Timo Aaltonen wrote: > On 21.2.2020 11.00, Hui Wang wrote: >> BugLink: http://bugs.launchpad.net/bugs/1861837 >> >> If the sof driver fails to load the firmware, it doesn't release the >> resource, then it make the system hang when suspending and resuming. >> >> After this patchset is applied, please also apply the patchset we >> just reverted 2 weeks ago, since this patchset is the fixing patches >> for the patches we reverted (#1857409). If need me to resend the >> reverted patches, I will resend them. >> >> >> [Impact] >> The machine hangs during suspend/resume if the sof driver fails to >> load firmware. >> >> [Fix] >> This bug is fixed in the mainline kernel v5.6-rc1, now backported >> those patches to ubuntu kernel >> >> [Test Case] >> sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot >> then suspend and resume, the system will not hang anymore >> >> And if the sof driver load the firmware successfully, the audio >> functions all worked well as before. >> >> [Regression Risk] >> Low, the patches are backported from upstream kernel, and I tested >> these patches on Lenovo and Dell dmic machines. >> >> >> Kai Vehmanen (2): >> ASoC: hdac_hda: Fix error in driver removal after failed probe >> ASoC: SOF: trace: fix unconditional free in trace release >> >> Pierre-Louis Bossart (2): >> ASoC: SOF: core: free trace on errors >> ASoC: SOF: core: release resources on errors in probe_continue >> >> Ranjani Sridharan (1): >> ASoC: SOF: Introduce state machine for FW boot >> >> sound/soc/codecs/hdac_hda.c | 4 +- >> sound/soc/sof/core.c | 86 ++++++++++++++++++++-------- >> sound/soc/sof/intel/hda-loader-skl.c | 1 - >> sound/soc/sof/intel/hda-loader.c | 1 - >> sound/soc/sof/intel/hda.c | 4 +- >> sound/soc/sof/ipc.c | 17 ++---- >> sound/soc/sof/loader.c | 19 ++++-- >> sound/soc/sof/pm.c | 25 +++++++- >> sound/soc/sof/sof-priv.h | 11 +++- >> sound/soc/sof/trace.c | 7 ++- >> 10 files changed, 125 insertions(+), 50 deletions(-) > applied to osp1 oem-next, thanks And after applying this patchset, we could reapply the patches we reverted before. (#1857409) If need me to resend the patches, just let me know. Thanks, Hui. >
On Fri, Feb 21, 2020 at 05:00:39PM +0800, Hui Wang wrote: > BugLink: http://bugs.launchpad.net/bugs/1861837 > > If the sof driver fails to load the firmware, it doesn't release the > resource, then it make the system hang when suspending and resuming. > > After this patchset is applied, please also apply the patchset we > just reverted 2 weeks ago, since this patchset is the fixing patches > for the patches we reverted (#1857409). If need me to resend the > reverted patches, I will resend them. > > > [Impact] > The machine hangs during suspend/resume if the sof driver fails to > load firmware. > > [Fix] > This bug is fixed in the mainline kernel v5.6-rc1, now backported > those patches to ubuntu kernel > > [Test Case] > sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot > then suspend and resume, the system will not hang anymore > > And if the sof driver load the firmware successfully, the audio > functions all worked well as before. > > [Regression Risk] > Low, the patches are backported from upstream kernel, and I tested > these patches on Lenovo and Dell dmic machines. Applied to focal/master-next, thanks!
On Tue, Feb 25, 2020 at 08:36:10PM -0600, Seth Forshee wrote: > On Fri, Feb 21, 2020 at 05:00:39PM +0800, Hui Wang wrote: > > BugLink: http://bugs.launchpad.net/bugs/1861837 > > > > If the sof driver fails to load the firmware, it doesn't release the > > resource, then it make the system hang when suspending and resuming. > > > > After this patchset is applied, please also apply the patchset we > > just reverted 2 weeks ago, since this patchset is the fixing patches > > for the patches we reverted (#1857409). If need me to resend the > > reverted patches, I will resend them. > > > > > > [Impact] > > The machine hangs during suspend/resume if the sof driver fails to > > load firmware. > > > > [Fix] > > This bug is fixed in the mainline kernel v5.6-rc1, now backported > > those patches to ubuntu kernel > > > > [Test Case] > > sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot > > then suspend and resume, the system will not hang anymore > > > > And if the sof driver load the firmware successfully, the audio > > functions all worked well as before. > > > > [Regression Risk] > > Low, the patches are backported from upstream kernel, and I tested > > these patches on Lenovo and Dell dmic machines. > > Applied to focal/master-next, thanks! Forgot to mention, I actually only applied patch 4 as 1-3 had already come in from upstream stable. Patch 1 at least was not completely identical to the upstream stable patch, so you may want to check to make sure it does what you need (I didn't really look into how they were different, possibly it is just comment/whitespace differences).
On 2020/2/26 上午10:42, Seth Forshee wrote: > On Tue, Feb 25, 2020 at 08:36:10PM -0600, Seth Forshee wrote: >> On Fri, Feb 21, 2020 at 05:00:39PM +0800, Hui Wang wrote: >>> BugLink: http://bugs.launchpad.net/bugs/1861837 >>> >>> If the sof driver fails to load the firmware, it doesn't release the >>> resource, then it make the system hang when suspending and resuming. >>> >>> After this patchset is applied, please also apply the patchset we >>> just reverted 2 weeks ago, since this patchset is the fixing patches >>> for the patches we reverted (#1857409). If need me to resend the >>> reverted patches, I will resend them. >>> >>> >>> [Impact] >>> The machine hangs during suspend/resume if the sof driver fails to >>> load firmware. >>> >>> [Fix] >>> This bug is fixed in the mainline kernel v5.6-rc1, now backported >>> those patches to ubuntu kernel >>> >>> [Test Case] >>> sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot >>> then suspend and resume, the system will not hang anymore >>> >>> And if the sof driver load the firmware successfully, the audio >>> functions all worked well as before. >>> >>> [Regression Risk] >>> Low, the patches are backported from upstream kernel, and I tested >>> these patches on Lenovo and Dell dmic machines. >> Applied to focal/master-next, thanks! > Forgot to mention, I actually only applied patch 4 as 1-3 had already > come in from upstream stable. Patch 1 at least was not completely > identical to the upstream stable patch, so you may want to check to make > sure it does what you need (I didn't really look into how they were > different, possibly it is just comment/whitespace differences). Checked, no functional difference. Thx. Hui.
On 21.02.20 10:00, Hui Wang wrote: > BugLink: http://bugs.launchpad.net/bugs/1861837 > > If the sof driver fails to load the firmware, it doesn't release the > resource, then it make the system hang when suspending and resuming. > > After this patchset is applied, please also apply the patchset we > just reverted 2 weeks ago, since this patchset is the fixing patches > for the patches we reverted (#1857409). If need me to resend the > reverted patches, I will resend them. > > > [Impact] > The machine hangs during suspend/resume if the sof driver fails to > load firmware. > > [Fix] > This bug is fixed in the mainline kernel v5.6-rc1, now backported > those patches to ubuntu kernel > > [Test Case] > sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot > then suspend and resume, the system will not hang anymore > > And if the sof driver load the firmware successfully, the audio > functions all worked well as before. > > [Regression Risk] > Low, the patches are backported from upstream kernel, and I tested > these patches on Lenovo and Dell dmic machines. > > > Kai Vehmanen (2): > ASoC: hdac_hda: Fix error in driver removal after failed probe > ASoC: SOF: trace: fix unconditional free in trace release > > Pierre-Louis Bossart (2): > ASoC: SOF: core: free trace on errors > ASoC: SOF: core: release resources on errors in probe_continue > > Ranjani Sridharan (1): > ASoC: SOF: Introduce state machine for FW boot > > sound/soc/codecs/hdac_hda.c | 4 +- > sound/soc/sof/core.c | 86 ++++++++++++++++++++-------- > sound/soc/sof/intel/hda-loader-skl.c | 1 - > sound/soc/sof/intel/hda-loader.c | 1 - > sound/soc/sof/intel/hda.c | 4 +- > sound/soc/sof/ipc.c | 17 ++---- > sound/soc/sof/loader.c | 19 ++++-- > sound/soc/sof/pm.c | 25 +++++++- > sound/soc/sof/sof-priv.h | 11 +++- > sound/soc/sof/trace.c | 7 ++- > 10 files changed, 125 insertions(+), 50 deletions(-) > For Eoan: Acked-by: Kleber Sacilotto de Souza <kleber.souza@canonical.com>
On 21.02.20 10:00, Hui Wang wrote: > BugLink: http://bugs.launchpad.net/bugs/1861837 > > If the sof driver fails to load the firmware, it doesn't release the > resource, then it make the system hang when suspending and resuming. > > After this patchset is applied, please also apply the patchset we > just reverted 2 weeks ago, since this patchset is the fixing patches > for the patches we reverted (#1857409). If need me to resend the > reverted patches, I will resend them. > > > [Impact] > The machine hangs during suspend/resume if the sof driver fails to > load firmware. > > [Fix] > This bug is fixed in the mainline kernel v5.6-rc1, now backported > those patches to ubuntu kernel > > [Test Case] > sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot > then suspend and resume, the system will not hang anymore > > And if the sof driver load the firmware successfully, the audio > functions all worked well as before. > > [Regression Risk] > Low, the patches are backported from upstream kernel, and I tested > these patches on Lenovo and Dell dmic machines. > > > Kai Vehmanen (2): > ASoC: hdac_hda: Fix error in driver removal after failed probe > ASoC: SOF: trace: fix unconditional free in trace release > > Pierre-Louis Bossart (2): > ASoC: SOF: core: free trace on errors > ASoC: SOF: core: release resources on errors in probe_continue > > Ranjani Sridharan (1): > ASoC: SOF: Introduce state machine for FW boot > > sound/soc/codecs/hdac_hda.c | 4 +- > sound/soc/sof/core.c | 86 ++++++++++++++++++++-------- > sound/soc/sof/intel/hda-loader-skl.c | 1 - > sound/soc/sof/intel/hda-loader.c | 1 - > sound/soc/sof/intel/hda.c | 4 +- > sound/soc/sof/ipc.c | 17 ++---- > sound/soc/sof/loader.c | 19 ++++-- > sound/soc/sof/pm.c | 25 +++++++- > sound/soc/sof/sof-priv.h | 11 +++- > sound/soc/sof/trace.c | 7 ++- > 10 files changed, 125 insertions(+), 50 deletions(-) > Acked-by: Stefan Bader <stefan.bader@canonical.com>
The bug status implies that this is NOT applied in Focal (despite the email) On 2020-02-21 17:00:39 , Hui Wang wrote: > BugLink: http://bugs.launchpad.net/bugs/1861837 > > If the sof driver fails to load the firmware, it doesn't release the > resource, then it make the system hang when suspending and resuming. > > After this patchset is applied, please also apply the patchset we > just reverted 2 weeks ago, since this patchset is the fixing patches > for the patches we reverted (#1857409). If need me to resend the > reverted patches, I will resend them. > > > [Impact] > The machine hangs during suspend/resume if the sof driver fails to > load firmware. > > [Fix] > This bug is fixed in the mainline kernel v5.6-rc1, now backported > those patches to ubuntu kernel > > [Test Case] > sudo mv /lib/firmware/intel/sof /lib/firmware/intel/sof-bak, reboot > then suspend and resume, the system will not hang anymore > > And if the sof driver load the firmware successfully, the audio > functions all worked well as before. > > [Regression Risk] > Low, the patches are backported from upstream kernel, and I tested > these patches on Lenovo and Dell dmic machines. > > > Kai Vehmanen (2): > ASoC: hdac_hda: Fix error in driver removal after failed probe > ASoC: SOF: trace: fix unconditional free in trace release > > Pierre-Louis Bossart (2): > ASoC: SOF: core: free trace on errors > ASoC: SOF: core: release resources on errors in probe_continue > > Ranjani Sridharan (1): > ASoC: SOF: Introduce state machine for FW boot > > sound/soc/codecs/hdac_hda.c | 4 +- > sound/soc/sof/core.c | 86 ++++++++++++++++++++-------- > sound/soc/sof/intel/hda-loader-skl.c | 1 - > sound/soc/sof/intel/hda-loader.c | 1 - > sound/soc/sof/intel/hda.c | 4 +- > sound/soc/sof/ipc.c | 17 ++---- > sound/soc/sof/loader.c | 19 ++++-- > sound/soc/sof/pm.c | 25 +++++++- > sound/soc/sof/sof-priv.h | 11 +++- > sound/soc/sof/trace.c | 7 ++- > 10 files changed, 125 insertions(+), 50 deletions(-) > > -- > 2.17.1 > > > -- > kernel-team mailing list > kernel-team@lists.ubuntu.com > https://lists.ubuntu.com/mailman/listinfo/kernel-team